Daniel Rall wrote:
>cmpilato@collab.net writes:
>
>>solo turn <soloturn99@yahoo.com> writes:
>>
>>>uhm ... i would be used to a compression level 0-9 or something like
>>>this. is there a reason why subversion wants to be different here?
>>>
>>Perhaps because we don't to specify a compression level? I was under
>>the impression that all this value does is tell neon to tell Apache
>>that, "Hey, if you feel like sending compressed responses, I'm cool
>>with that." Does neon *send* compressed requests, too?
>>
>
>I cannot say definitively whether it does or does not, but when I was
>using ethereal to monitor the communication between HTTP client and
>server, I never saw neon send a compress request. Rather, the client
>sends the Accept-Encoding HTTP header with a value of "gzip",
>indicating to the server that it finds only gzip-encoded responses
>acceptable. As HTTP is a stateless protocol and servers are not (to
>my knowledge) required to support gzip, it wouldn't really make sense
>for the client to expect that the server will be able to understand
>gzip.
>
>Section 14.3 of RFC 2616 provides a more thorough explanation:
>http://www.w3.org/Protocols/rfc2616/rfc2616-sec14.html#sec14.3
>
The client can discover if the server supports it by issueing an OPTIONS
request.
You typically only need to do this once per session. Unless the admin
disables
compression somewhere in the URI space below the repository root.
Sander
---------------------------------------------------------------------
To unsubscribe, e-mail: dev-unsubscribe@subversion.tigris.org
For additional commands, e-mail: dev-help@subversion.tigris.org
Received on Thu Nov 21 23:55:27 2002